What is claude-mcp-bridge AGENTS.md?

Repository-specific instructions for claude-mcp-bridge, an open-source program that lets coding tools call Claude Code through the Model Context Protocol. They describe the project’s structure, tools, security settings, and development rules.

Use them when modifying the TypeScript MCP server, its Claude CLI subprocess handling, structured tools, prompts, or authentication and isolation behavior.

They give coding agents the project context and constraints needed to work consistently in this repository. They also explain how authentication and subprocess isolation affect changes.

How hampsterx/claude-mcp-bridge configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to hampsterx/claude-mcp-bridge. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/hampsterx/claude-mcp-bridge/master/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/hampsterx/claude-mcp-bridge

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/hampsterx/claude-mcp-bridge/agents-md/github.svg)](https://agentmods.dev/instructions/hampsterx/claude-mcp-bridge/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/hampsterx/claude-mcp-bridge/agents-md"><img src="https://agentmods.dev/badge/instructions/hampsterx/claude-mcp-bridge/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/hampsterx/claude-mcp-bridge/agents-md"><img src="https://agentmods.dev/badge/instructions/hampsterx/claude-mcp-bridge/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,487 This file is loaded in full into every session.

When invoked 2,487 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.02487	$0.02487
Opus 5.5	$0.00995	$0.00995
Sonnet 5.5	$0.00497	$0.00497
Haiku 4.5	$0.00249	$0.00249

Measured 2mo ago against content hash 47012fc2a354, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

claude-mcp-bridge AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 171 lines

How it starts

The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md - claude-mcp-bridge

Guidance for AI coding agents working in the claude-mcp-bridge repository.

This file defines repository-specific operating rules for autonomous or semi-autonomous coding agents. Follow these instructions unless a maintainer explicitly tells you otherwise.

Project Overview

Open source MCP server that wraps Claude Code CLI as a subprocess, exposing code execution, web search, and structured output as MCP tools. Works with any MCP-compatible client: Codex CLI, Gemini CLI, Cursor, Windsurf, VS Code.

npm package: claude-mcp-bridge
License: MIT
Language: TypeScript
Framework: @modelcontextprotocol/sdk

Architecture

MCP Client  --stdio-->  claude-mcp-bridge  --spawn-->  claude CLI subprocess

Prompts are assembled in TypeScript and spawned via the Claude Code CLI. Auth determines the spawn flags: API-key auth (CLAUDE_BRIDGE_USE_API_KEY=1) uses --bare for maximum isolation (skips hooks, memory, plugins, CLAUDE.md loading); subscription auth (the default) runs non-bare so the CLI can resolve OAuth tokens, with --setting-sources "" preventing project settings from leaking into the subprocess. See DESIGN.md § Subprocess Spawning and SECURITY.md § Isolation by Auth Mode. The search tool loads its prompt template from prompts/*.md via src/utils/prompts.ts and fills placeholders.

Code review is not a bridge tool. Use Claude Code's built-in /review family in-session, or invoke claude -p directly with the hardened isolation flags documented in README § Code review with this CLI. Rationale: ADR-001.

Tools

Tool	Purpose	Default Timeout
`query`	Execute prompts with file context, session resume, budget control	60s (text) / 120s (images)
`search`	Web search via Claude CLI WebSearch/WebFetch tools	120s
`structured`	JSON Schema validated output via `--json-schema`	60s
`listSessions`	List active sessions with cumulative cost and turn counts	instant
`ping`	Health check + CLI capability detection	10s

Read the full file on GitHub · 171 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 171 lines · 2,487 tokens per session scan A 47012fc2a354

Subscribe to this mod's changes

claude-mcp-bridge AGENTS.md is an instructions file published in the GitHub repository hampsterx/claude-mcp-bridge (7 stars, last pushed 19d ago), licensed MIT. It adds 2,487 tokens to every session, about $0.0099 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents

claude-mcp-bridge AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
claude-mcp-bridge AGENTS.mdhampsterx/claude-mcp-bridge	Instructions file	7	19d ago	MIT	A	2,487	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	275k	2d ago	MIT	A	4,313	Codex, OpenCode

claude-mcp-bridge: Instructions file for Codex